305 



Since the 'N.W. passage of the wind is the critical time for rain, a know- 

 ledge of the wind's days enables ns to forecast the days Avhich will be precarious 

 in this respect. In bad weather, there are often two revolutions in one true 

 period, and then there are two critical days in the week — the secondary often 

 being greater than the primary. At other times, the wind remains steady at 

 one point for a whole period, or even two, the crisis being marked by a few 

 hours of calm. 



The exact determination of this period over long intervals is a problem of 

 great interest, as it would help to throw light on the causes .which determine 

 the period, whether it has any connection with the phases of the moon, as has 

 often been conjectured, or is the result of a certain equilibrium between the 

 strength of the polar and equatorial currents and the earth's axial motion, or 

 other caiises. 



The characteristics of the wind in the different quarters are well marked 

 and constant. The warm IST., the stormy and rainy N.W. and W., the cold 

 damp S., and the calm dry E., are invariable. The barometer also invariably 

 rises with a S. and E., and falls with a N. and W. wind. The manner in 

 which the various winds displace each other is also a point of scientific interest. 

 Professor Dove found that the S. wind was blowing in the iipper regions before 

 it descended to the surface in the Northern Hemisphere ; or, in other words, 

 that the IS", wind there displaces the S. wind first in a thin stratum at the 

 surface, acquiring depth as it advances. He found that at the Peak of 

 TenerifFe it is often blowing a S.W. wind when the N.E. wind is blowing at 

 the base. The corresponding case seems to be true for the Soiithern Hemis- 

 phere, since at elevated stations, such as Bealey, the wind is often N'.W., 

 while the S. wind has set in at lower levels. The barometer also gives similar 

 evidence, as it rises only when the S. wind actually sets it, while it falls for 

 some time previous to the advent of the N. wind at the surface. 



The occurrence of oiir white frosts before a N. wind and rain, would thus 

 receive an explanation, as being the result of the warm dry N. wind in the 

 higher regions bearing down towards the surface on the strata of cold damp 

 S. wind, and by its evaporating power changing the vapour from a grosser 

 to a finer state of particle, and hence causing cold in the regions below from a 

 well known law, the intensity of the cold becoming a measure of the -rapidity 

 of evaporation and the approach of complete saturation or rain. 



Another interesting point is the bearings of the line of displacement. This 

 could only be determined with any accitracy by a combined system of simul- 

 taneous observations at distant stations free from much local disturbance. The 

 form of the advancing area within which the S. wind is blowing, and the 

 velocity with which it travels, are also points of great importance as indicating 

 the probable path of cyclonic disturbance and the rate of their advance, both 

 of which ai-e found to be totally independent of the direction and force of the 



2 R. 



